I put in the automatic watering timer on the primary watering network (vegetables), and a manual timer on the secondary watering network (fruits) last night, and let them do their thing automatically this morning.

When I came home in the evening, I found a section of hose had popped out from the second system, next to the pear tree; most likely the initial water pressure was too much and it came right off. So a spot on the grass got some nice water for two hours.

I also discovered that there is a limitation to adding more extensions to a watering system.  Basically with each new extension (a split somewhere), the time required to water all the beds plus new extensions increases. I have not determined yet but with five different splits, the primary watering system needs three hours to get everything. Likewise I am limited by the amount of pressure my hand-made system can tolerate. If I open the faucet above a certain point, the watering system comes apart from pressure. So I have to water everything longer, instead of water everything with more water in the same time period.

The secondary system needs about two hours (provided it doesn’t burst apart again). I re-attached the parts that popped off, twice.

A day of watering, measured

Three hours on the primary watering system consumed 950 liters, or 250 gallons. The timer is set to do this every two days.

Update: I measured the second watering system (the “fruit watering system” I call it), which waters at 5.5 liters per minute. That’s 660 liters (174 gallons) in two hours. I should reduce the output a bit since it waters much less area than the vegetable garden (one fruit tree and five fruits).

Future estimates

Assuming I water every two days, from today, July 16 to October 1st, (77 days in between), that’s 950 liters on 38 days, a projected water consumption of 36,100 liters (9,536 gallons).

The secondary system at the current flow rate will use 25,100 liters (6630 gallons). I estimated it was 1/3 of the primary but it’s closer about 2/3 of the primary system’s usage.

In total, a season’s water use could be up to 61,200 liters (16,160 gallons). At my city’s rates ($5.41 per thousand gallons) that will cost about $87 dollars.

61,200 liters (16K gallons) still seems like a lot

At this point, $87 is a drop in the bucket. I could easily afford to run the watering system every day (only doubling the costs over the season to ~$180). But that would take more water from the city.

It seems more ecologically friendly to harvest rainwater

Why tap the fresh water supply when rain catchment could offset that? I did some rough calculations on how much water my roof could collect in the year.

Maryland’s average yearly rainfall is 41 inches.

My 1,500 square-foot roof catches 216,000 cubic inches of water per one inch of rainfall.

With 41 inches of rain a year my roof would catch 41 x 216,000 = 8,856,000 cubic in/year.

231 cubic inches per gallon makes 38,337 gallons captured annually.

My naive estimates conclude that a water catchment system would surely catch enough water to supply the garden year-round.

How many gallons of storage would I need?

From this website, https://www.currentresults.com/Weather/Maryland/average-yearly-precipitation.php, there are 107 days of precipitation in Frederick. If it rains when the temperature is above freezing, in Frederick MD, 9 months of the year’s precipitation would be rain. The rest will be snow.

If in 12 months there are 107 precipitation days, in 9 above-freezing months there are 80.25 precipitation days. Nearly every 9th day (8.9) is a rain day. See this link for temperatures: http://www.intellicast.com/Local/History.aspx?month=2

If in a year it rains 41 inches, each precipitation day, averaged, is 0.51 inches per rain day. On my roof which captures 216K cubic inches per rainfall-inch, on an average rain day it will collect 110,061 cubic inches (476 gallons).

If three hours on the primary system and two hours on the secondary system consume in total (an estimated) 420 gallons per watering run, a rain day will deliver 1.13 “automated watering days.”

On days which it rains, the automated system will not be needed, so I will not run it. The theoretical water harvesting system will collect water.

A scenario in which it rains every 9 days:

Day 1: Rain day. No automated watering needed. +476 gallons stored
Day 2: No automated watering, since it just rained.
Day 3: Automated watering. -420 stored gallons used (56 stored remaining)
Day 4: No watering
Day 5: Automated watering. -420 gallons (0 stored remaining, 364 taken from city)
Day 6: No watering
Day 7: Automated watering. -420 gallons from city
Day 8: No watering
(Day 9: Rain day)

So in the 9-day span,  ~784 gallons taken from the city, 476 stored gallons used.

Over 77 days left in the season (assuming my stored watering system appears right now, fully built) that ends up being:

Every 9 days, a theoretical water harvesting system assisting my current water usage takes 784 gallons from the city.

77 days divided by 9 rain periods makes 8.5 stretches of rain. So 8.5 * 784 makes 6,660 gallons taken from the city.

Without any theoretical rainwater harvesting, that would be 1260 every 9 days (accounting for the rain day), times 8.5 makes 10,710 gallons taken from the city.

Rainwater harvesting would save 37% more water.

Well, something doesn’t add up, if there are 38K gallons of precipitation on my roof annually, why am I still taking water from the city? Is some of it snow?

Is a 476-gallon system good enough or should I account for thunderstorms and excess water, maybe set something up that’s 750 gallons?

Can I harvest snow and store it as water for the summer?

Interesting (and improved) calculations ahead!

To calculate:

• The total growing season, and the number of rain days therein,
• The more accurate amount of rain that can be captured, annually, accounting for snow
• Better run rate vs storage rate
• How much actual water storage would I need to 100% capture all precipitation, and use it on the garden.